floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

scal benchmark: eliminate y, move init/timing out of loop 

Removing y avoids cache effects (if y is the size of the L1 cache, the
main array x is removed from it).
Moving init and timing out of the loop makes the scal benchmark behave like
the gemm benchmark, and allows higher accuracy for smaller test cases since
the loop overhead is much smaller than the timing overhead.

Example:
OPENBLAS_LOOPS=10000 ./dscal.goto 1024 8192 1024
on AMD Zen2 (7532) with 32k (4k doubles) L1 cache per core.

Before
From : 1024  To : 8192 Step = 1024 Inc_x = 1 Inc_y = 1 Loops = 10000
   SIZE       Flops
   1024 :     5627.08 MFlops   0.000000 sec
   2048 :     5907.34 MFlops   0.000000 sec
   3072 :     5553.30 MFlops   0.000001 sec
   4096 :     5446.38 MFlops   0.000001 sec
   5120 :     5504.61 MFlops   0.000001 sec
   6144 :     5501.80 MFlops   0.000001 sec
   7168 :     5547.43 MFlops   0.000001 sec
   8192 :     5548.46 MFlops   0.000001 sec

After
From : 1024  To : 8192 Step = 1024 Inc_x = 1 Inc_y = 1 Loops = 10000
   SIZE       Flops
   1024 :     6310.28 MFlops   0.000000 sec
   2048 :     6396.29 MFlops   0.000000 sec
   3072 :     6439.14 MFlops   0.000000 sec
   4096 :     6327.14 MFlops   0.000001 sec
   5120 :     5628.24 MFlops   0.000001 sec
   6144 :     5616.41 MFlops   0.000001 sec
   7168 :     5553.13 MFlops   0.000001 sec
   8192 :     5600.88 MFlops   0.000001 sec

We can see the L1->L2 switchover point is now where it should be, and the
number of flops for L1 is more accurate.
This commit is contained in:
Bart Oldeman 2022-11-29 08:02:45 -05:00
parent e9a911fb9f
commit bae45d94d1
1 changed files with 9 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
benchmark/scal.c
View File

@ -45,7 +45,7 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
int main(int argc, char *argv[]){
FLOAT *x, *y;
FLOAT *x;
FLOAT alpha[2] = { 2.0, 2.0 };
blasint m, i;
blasint inc_x=1,inc_y=1;
@ -74,10 +74,6 @@ int main(int argc, char *argv[]){
fprintf(stderr,"Out of Memory!!\n");exit(1);
}
if (( y = (FLOAT *)malloc(sizeof(FLOAT) * to * abs(inc_y) * COMPSIZE)) == NULL){
fprintf(stderr,"Out of Memory!!\n");exit(1);
}
#ifdef __linux
srandom(getpid());
#endif
@ -91,30 +87,20 @@ int main(int argc, char *argv[]){
fprintf(stderr, " %6d : ", (int)m);
for(i = 0; i < m * COMPSIZE * abs(inc_x); i++){
x[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
}
begin();
for (l=0; l<loops; l++)
{
for(i = 0; i < m * COMPSIZE * abs(inc_x); i++){
x[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
}
for(i = 0; i < m * COMPSIZE * abs(inc_y); i++){
y[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
}
begin();
SCAL (&m, alpha, x, &inc_x);
}
end();
end();
time1 = getsec();
time1 = getsec();
timeg += time1;
}
timeg /= loops;
timeg = time1 / loops;
#ifdef COMPLEX
fprintf(stderr, " %10.2f MFlops %10.6f sec\n", 6. * (double)m / timeg * 1.e-6, timeg);